I spotted this earlier today whilst searching for something else. It was published in 2021:
Roadmap To Becoming A Test Automation Engineer
Fascinating to reflect on what can happen in 4 years.
Have a read and let others know, what’s changed since then? What seem to be familiar themes that haven’t changed over time?
What would a future roadmap to becoming an automation engineer look like?
AI in Testing / Automation can be added along with mention of tools like GitHub Copilot and other AI assistant be it ChatGPT.
Outdated tools like TestProject can be removed.
But basics remains same and its well articulated how having a mentor can help someone starting with a journey
One thing that has changed in last 4 year is introduction of Gen-AI based tools and their importance in automation.
Even though the article cover in detail the roadmap to become automation engineer but if we talk about the roadmap in today’s scenario then it should include familarity with tools like Copilot cursor etc. and how someone can use them to leverage the poweer of automation in testing.
Apart from that since the article was published in 2021 the other major change is importance of Playwright in the market.
Last 4 years playwright has evolved as the direct competitor of Selenium and many mid-small companies are adopting it for automation.
So if someone wants to become automation engineer then their roadmap can be with selenium as well as with playwright also.
However at the end of days the fundamentals remain same the things that shape the automation remain same.
API Automation is also now very much demand in the market along with UI and mobile app automation.
So these are the some key changes that we can notice in last 4 years for someone who is looking for roadmap to become automation engineer.
The article covered quite a broad balanced approach but perhaps a bit optimistic compared to what a lot of automators focused on.
Do you actually need a strong foundation in testing, has that need diminished even more with ai and other low code tools?
What a lot of automation does is focus on the known risk side of things, the elements that can be scripted. This is a significant subset of testing but its is a model of testing that many companies follows so do you need strong foundations in testing as a whole or primarily this smaller subset of testing?
AI can significantly fast track that roadmap in my current view so perhaps similar roadmap but shorter journey and maybe even more rounded scripts as a result.
Whether that though compromises actual understanding is a risk but at this point will the need for rapid churn outweigh that need, particularly given a lot of the older paths were often also just towards the script churn goals rather than architecture level skills.
It’s a different consideration but Will AI amplify the risks, faster churn, opportunity cost value comparison of doing deeper testing to the extent the latter gets dropped even more and the quality bar drops. Will this be recognised and accepted, risking changing the bigger picture testing roadmap on the way.
The updated roadmaps may give us some early indications of the paths ahead not just for automation engineers but for testing as a whole.
I think a lot of the article sticks.
I agree with @andrewkelly2555 that I don’t think there is a need for a test automation engineer to have strong foundation in testing specifically…but someone has to.
The vast majority of automation engineers I’ve worked were dedicated automation engineers, they loved coding and didn’t want to get involved in requirement analysis, design and manual test execution. Their expertise was analysing the product and its architecture and coming up with an automation strategy that would suit the product…but not necessarily what was the most important thing to automate. So its interesting that the article referenced when building an automation strategy that the one resource they didn’t mention collaborating with was other testers - who often have all the answers to those missing pieces.
I see a few comments around AI and sure AI can help with automation and wasn’t so prominent then. But I would question whether its a tool to actually “learn” automation. You need to have enough skills to know when AI comes up with the wrong answer and not fall into the trap of trying to make AI give you the right answer.
I do think one thing that is becoming more common, is the building utilities to aid testability. So thats where automation engineers skill sets are being extended closer to developers beyond just testing UI’s, API’s and databases - but tools to bridge those testability gaps. The need for CI/CD hasn’t diminished over the years so I think thats still just as prominent.